Normalization of gearbox vibration signal for tooth crack diagnosis under variable speed conditions

Xingkai Yang,Peng Zhou,Ming J. Zuo,Zhigang Tian

DOI: https://doi.org/10.1002/qre.3029

2021-12-03

Quality and Reliability Engineering International

Abstract:Variable speed conditions introduce Amplitude Modulation (AM) and Frequency Modulation (FM) effects into gearbox vibration signals, which makes it difficult to distinguish between changes of tooth crack severity and speed changes. To overcome this problem, the AM and FM effects caused by speed variation need to be removed. Order tracking techniques are used to remove the FM effect. Some methods have been reported to reduce the AM effect. However, they attenuated crack information since they focused on the entire vibration signal. Besides, the performance of the reported methods on removing the AM effect was not quantitatively evaluated. In this study, a novel normalization method focusing on the Crack Induced Impulses (CII) is proposed to remove the AM effect without attenuating the tooth crack information. A modified Adaptive Chirp Mode Decomposition method is developed to obtain the CII under variable speed conditions. The peak envelope of the CII is determined using spline interpolation of its envelope peaks and is employed to remove the AM effect of the CII by normalization. Two metrics are introduced to quantitatively evaluate the performance of the proposed normalization method on removing the AM effect and preserving the tooth crack information. The effectiveness of the proposed normalization method is demonstrated using simulated gearbox signals and experimental gearbox datasets. The proposed method benefits tracking tooth crack severity progression under variable speed conditions.

engineering, industrial, multidisciplinary,operations research & management science